![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to sn0 CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Research Unix] / researchv10no / cmd / f2c / doc|
Return to sn0 CVS log | Up to [Research Unix] / researchv10no / cmd / f2c / doc |
1.1 root 1: .EQ
2: define dollar % "\f(CW$\fP" %
3: delim $$
4: define f2c % "f\|2c" %
5: define F2c % "F\^2c" %
6: define libF77 % "libF77" %
7: define libI77 % "libI77" %
8: define LibF77 % "LibF77" %
9: define LibI77 % "LibI77" %
10: .EN
11: .TL
12: Availability of $f2c$ \(em a Fortran to C Converter
13: .AU
14: S. I. Feldman
15: .AI
16: Bellcore
17: Morristown, NJ 07960
18: .AU
19: D. M. Gay
20: .AI
21: .MH
22: .AU
23: M. W. Maimone
24: .AI
25: Carnegie-Mellon University
26: Pittsburgh, PA 15213
27: .AU
28: N. L. Schryer
29: .AI
30: .MH
31: .AB \
32: .AE
33: .PP
34: Automatic conversion of Fortran 77
35: .[ [
36: ANSI FORTRAN 1978
37: .]]
38: to C
39: .[ [
40: Kernighan Ritchie 1988
41: .]]
42: is desirable for
43: several reasons. Sometimes it is useful to run a
44: well-tested Fortran program on a machine that has a C
45: compiler but no Fortran compiler. At other times, it
46: is convenient to mix C and Fortran. Some things are
47: impossible to express in Fortran 77 or are harder
48: to express in Fortran than in C
49: (e.g. storage management, some character operations,
50: arrays of functions, heterogeneous data structures,
51: and calls that depend on the operating system),
52: and some programmers simply prefer C to Fortran.
53: There is a large body of well tested
54: Fortran source code for carrying out a wide variety of
55: useful calculations, and it is sometimes desirable to
56: exploit some of this Fortran source in a C environment.
57: Many vendors provide some way of mixing C and Fortran, but
58: the details vary from system to system.
59: Automatic Fortran to C conversion lets one create a
60: .I portable
61: C program that exploits Fortran source code.
62: .PP
63: A side benefit of automatic Fortran 77 to C conversion is that
64: it allows tools like
65: .I lint
66: \
67: .[[
68: v101
69: .]]
70: to provide Fortran 77 programs with some of the consistency
71: and portability checks that the Pfort Verifier
72: .[ [
73: Ryder 1974
74: .]]
75: made available to Fortran 66 programs.
76: .PP
77: Starting from Feldman's original $f77$ compiler
78: .[ [
79: Feldman Weinberger Portable Fortran
80: .]]
81: (the original UNIX\u\(rg\d Fortran 77 compiler),
82: we have created a program called $f2c$ that converts
83: Fortran 77 source code into C.
84: Our report
85: .[ [
86: Feldman Gay Maimone Schryer
87: .]]
88: describes $f2c$'s conversions in considerable detail;
89: the purpose of this note is to point out the availability
90: both of our report and of the source for $f2c$ itself.
91: On request, we are happy to send out paper copies of the report.
92: You can also obtain a Postscript version of it by electronic
93: mail by sending the message
94: .ce
95: \f(CWsend f2c.ps from f2c\fR
96: to \f([email protected]\fR. Source for both $f2c$
97: and the support libraries assumed by the C it produces
98: is available by electronic mail and by $ftp$. For details,
99: send the electronic-mail message
100: .ce
101: \f(CWsend index from f2c\fR
102: to \f([email protected]\fR. You can also $ftp$ to
103: \f(CWresearch.att.com\fR; in subdirectory \f(CWdist/f2c\fR you
104: will find a copy of the \f(CWindex\fR file and of source for
105: $f2c$ and its support libraries.
106: (As of this writing, 426 different people had
107: obtained $f2c$ source from
108: .I netlib ,
109: i.e., by electronic mail,
110: and 644 had obtained it by $ftp$.)
111: .PP
112: We have used $f2c$ to convert various large programs and
113: subroutine libraries to C automatically (i.e., with no manual intervention);
114: these include the \s-2PORT3\s+2 subroutine library (\s-2PORT1\s+2
115: is described in
116: .[ [
117: Fox Hall Schryer Algorithm 1978
118: .]
119: .[
120: Fox Hall Schryer port 1978
121: .]]),
122: MINOS
123: .[ [
124: Murtagh Saunders 1987
125: .]],
126: and Schryer's floating-point test
127: .[ [
128: Schryer floating
129: .]].
130: In addition to our own testing, we have been helped by feedback
131: from many people and by observing $f2c$'s behavior on some 21 megabytes
132: of Fortran submitted to $netlib$'s experimental ``\f(CWexecute f2c\fR''
133: service; the feedback and our observations have led us to isolate
134: and repair numerous bugs and to extend $f2c$ in various ways.
135: .PP
136: Although we have tried to make $f2c$'s output reasonably readable,
137: our goal of strict compatibility with $f77$ implies some nasty
138: looking conversions. Input/output statements, in particular,
139: generally get expanded into
140: a series of calls on routines in $libI77$ ($f77$'s I/O library).
141: Thus the C output of $f2c$ would probably be something of a nightmare
142: to maintain as C; it is much more sensible to maintain the original
143: Fortran, translating it anew each time it changes. Some commercial
144: vendors, e.g., those listed in an appendix to our report,
145: seek to perform translations yielding C that one
146: might reasonably maintain directly; in general, these translations
147: require some manual intervention.
148: .PP
149: $F2c$ optionally emits special files called
150: .I prototype
151: files that summarize the calling sequences of
152: the translated subprograms.
153: A side benefit of $f2c$ is that it can read prototype
154: files and check the consistency of calling sequences
155: across files. We have found this to be a useful debugging aid.
156: Use of prototypes is slightly more convenient than invoking
157: .I lint ,
158: but
159: .I lint
160: warns of other possible errors, such as variables that may be
161: referenced before they are set.
162:
163: .ce
164: REFERENCES
165: .LP
166: .so tmac.sdisp1
167: .[
168: $LIST$
169: .]
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.